|Topic:||Question about importing protein/peptide list (Maxquant-based library) .|
|Details:|| First I want to thank you for having this nice forum set up. I really benefit a lot from reading the existing discussions to get more familiar with the use of the software. And I appreciate your time and effort spent in maintaining this robust trouble shooting environment and offering kind help.
My question is related to one simple concept: library construction. It might sound a little weak, but I really want to figure out the basics behind this. We all know that building a library only takes a few seconds, but I'm puzzled by the difference between two ways of importing your targeted protein/peptide list out of your library: 1)Go to the spectral library then add all the peptides with associated proteins; 2)Import a FASTA file directly to pick all the library-matching peptides/proteins. Do they serve for different purposes?
My second question is somewhat related to the first one. If I chose to import all peptides with associated proteins from a specific library into the document, I would only see 1/10 of the target proteins recovered (200 proteins compared to my Maxquant search results of 2000 proteins). I understand the concept of redundant and non-redundant library, but I'm just confused about why I lost so many of the proteins in my result. Is there a way to put a less strict cut-off so that I can import all targets?
Posted by: adr
2017-11-14 21:49:42 MST
| On the first question, I think you have outlined two ways of achieving much the same result, but the two features do provide different options that adds flexibility to the problem of getting targets into Skyline for further analysis. The first requires that you set up a background proteome first, while the second does not. The second also allows you to create FASTA outside Skyline that describes a subset of proteins you are interested in targeting, which may have been derived completely separately from what was detected in your samples. With very large searches, I think the FASTA import option may also provide better feedback and quicker response. The second obviously is nicely connected with viewing the underlying data in the Spectral Library Explorer. But, yes, you could fairly easily construct a set of steps where 1 and 2 produce the same end result.
On the second question, you can certainly change your cut-off when you build the spectral library. You could use 0.99, 0.95, 0.5 or even 0.0 (for 1%, 5%, 50% and unrestricted FDR). We don't really recommend using 0 unless you know that your results are pre-filtered and contain only the peptide spectrum match results you want in your library. When you look in the Spectral Library Explorer does it look like the number of peptides is what you expect or is that also greatly reduced from what MaxQuant is telling you for your desired FDR?
COPYRIGHT ©2019 . All Rights Reserved.